Transgenic and targeted gene disruptions afford powerful animal models for the characterization of toxicant-induced effects in murine models. Three basic approaches will be used in this collaborative R01 grant to develop murine transgenic models for the characterization of toxicant actions. In the first project, transgenic mice will be developed in which reporter genes will be inserted under control of glial fibrillary acidic protein (GFAP) promoter elements. Since elevations in GFAP can be used to identify brain regions damaged by neurotoxicants, coupling elements which regulate GFAP expression to the Lac-Z gene and the Green Fluorescent Protein (GFP) will yield mice which are predicted to show either colorimetric or fluorometric changes in GFAP expression following toxicant administration. The initial Lac-Z-GFAP mice have been established as a founder line. Such mice should be powerful predictors of know and suspected neurotoxicants. Mice with GFAP-GFP transgenes will be bred with lines contatining a deletion of inducible nitric oxide synthase (iNOS), and homozygous progeny used to test the hypothesis that specific excitotoxicants will produce less neurotoxicity as measured by reporter genes. In the second project a transgenic targeted cellular ablation model will be used to produce mice in which Purkinje cells can be selectively lesioned in adults. In this model, the viral thymidine kinase transgene is inserted under control of the L7 Purkinje cell specific promoter; after maturation, gancyclovir is administered, producing cellular toxicity only in cells expressing the transgene. The advantage of this model system is that cell-specific promoters can be used to conditionally affect specific neuronal populations after development. This approach was used to produce selective parietal cell ablations. Purkinje cell deficient mice will then be used to assess the effects of cerebellar toxicants such as harmaline and 3-acetylpyridine on neuronal function and histology. In the third project normal and mutated forms of the important DNA repair enzyme 06-alkylguanine-DAN alkyltransferase (AGT) will be expressed in the mouse lung using promoters specific for either type II or Clara cells. Mice which are homozygous for either Clara cell or type II cell expression of AGT will then be used to determine whether the presence of this repair enzyme protects from the effects of environmental toxicants such as 4- (methylnitrosamine)-1-(3-pyridy1)-1-butanone (NKK) and related compounds, to investigate the cell types in which tumors originate, and to test the nature of the critical adducts formed in DNA